kubis.ai USE AI NOW, ASK ME HOW
Interface Theme
Typography Set
Layout Mode
Ouroboros Loop
  • Timeline
  • Plain MD

Posts

  • Blog (35)
    • AI Augmented Workflow 10
      • Claude Code Skills 5
      • Vscode Githubcopilot 5
    • Dgx_series 16
      • Benchmarks 4
    • Llms 5
      • Tokens In Logits Out 5
    • Markdown Et Al 3
  • Archive
    • 2026 21
      • April 2
      • March 19
    • 2025 14
      • December 8
      • August 5
      • June 1
  • Tags
    • AGENTS.md 1
    • AI 29
    • API 1
    • Agent SDK 1
    • BPE 1
    • Blackwell 1
    • CUTLASS 1
    • Claude Code 8
    • DGX Spark 16
    • DevOps 1
    • GPT-2 5
    • GPT-OSS-120B 2
    • Gemma4 2
    • Git 1
    • GitHub Copilot 4
    • Groq 1
    • Hugging Face 1
    • LLM 5
    • LiteLLM 4
    • Local AI 16
    • Marp 2
    • Mermaid 2
    • Nemotron 2
    • Ollama 1
    • Open WebUI 1
    • OpenAI 1
    • Ray 1
    • SentencePiece 1
    • VSCode 5
    • agents 1
    • ai-augmented-workflow 10
    • attention 1
    • automation 1
    • benchmarking 7
    • btop 1
    • cluster 2
    • diagrams 1
    • documentation 1
    • finance 1
    • introduction 1
    • llama-benchy 1
    • markdown 4
    • markdown-et-al 3
    • marp 1
    • memory bandwidth 1
    • mermaid 1
    • monitoring 1
    • no-code 1
    • plugins 1
    • presentations 2
    • proxy 1
    • quantization 1
    • recipes 1
    • reconciliation 1
    • sampling 1
    • skills 5
    • temperature 1
    • tokenization 1
    • top-k 1
    • top-p 1
    • transformer 2
    • tutorial 1
    • vLLM 11
    • visualization 1

Posts tagged with "transformer"

Found 2 posts

August 24, 2025 · 6 min read

3/5 The Transformer in 90 Seconds (Then the Other 900)

Attention is a weighted mix. Multi-head is a filter bank. The causal mask means no spoilers. Here's the transformer architecture without the math — then with just enough of it.

#LLM #transformer #attention #GPT-2 #AI
August 18, 2025 · 4 min read

1/5 Tokens In, Logits Out: What's Actually Inside ChatGPT

You use ChatGPT, Gemini, Claude every day. But what's inside the box? Assistants are not models. Once you see the difference, you unlock controls most people don't know exist.

#LLM #GPT-2 #transformer #AI
← Back to all posts
Sidebar
Pin sidebar ›
Hub Timeline Archives
  • Timeline
  • Plain MD

Posts

  • Blog (35)
    • AI Augmented Workflow 10
      • Claude Code Skills 5
      • Vscode Githubcopilot 5
    • Dgx_series 16
      • Benchmarks 4
    • Llms 5
      • Tokens In Logits Out 5
    • Markdown Et Al 3
  • Archive
    • 2026 21
      • April 2
      • March 19
    • 2025 14
      • December 8
      • August 5
      • June 1
  • Tags
    • AGENTS.md 1
    • AI 29
    • API 1
    • Agent SDK 1
    • BPE 1
    • Blackwell 1
    • CUTLASS 1
    • Claude Code 8
    • DGX Spark 16
    • DevOps 1
    • GPT-2 5
    • GPT-OSS-120B 2
    • Gemma4 2
    • Git 1
    • GitHub Copilot 4
    • Groq 1
    • Hugging Face 1
    • LLM 5
    • LiteLLM 4
    • Local AI 16
    • Marp 2
    • Mermaid 2
    • Nemotron 2
    • Ollama 1
    • Open WebUI 1
    • OpenAI 1
    • Ray 1
    • SentencePiece 1
    • VSCode 5
    • agents 1
    • ai-augmented-workflow 10
    • attention 1
    • automation 1
    • benchmarking 7
    • btop 1
    • cluster 2
    • diagrams 1
    • documentation 1
    • finance 1
    • introduction 1
    • llama-benchy 1
    • markdown 4
    • markdown-et-al 3
    • marp 1
    • memory bandwidth 1
    • mermaid 1
    • monitoring 1
    • no-code 1
    • plugins 1
    • presentations 2
    • proxy 1
    • quantization 1
    • recipes 1
    • reconciliation 1
    • sampling 1
    • skills 5
    • temperature 1
    • tokenization 1
    • top-k 1
    • top-p 1
    • transformer 2
    • tutorial 1
    • vLLM 11
    • visualization 1

April 9, 2026 ·